I have a simulated a SC repeated dosing PK profile. The model is a one compartmental model with 1 order absorption. I have between subjects variation on all 4 parameters (ka, Cl, Vd and F). I would like a prediction with no model variation, just the between subject variation.
The average profile of 1000 simulations looks like Picture 1 (left)
It is a bit noise and a found out that its an average of 1000 very noise single profiles Picture 2 (right)
The Ceps is set to 0
Should a single profile with only between subject variation not be without variation/noise?
Is there a way to eliminate the noise? I tried (increasing/reducing the no of time points with no luck)
Are the sampling times the same for each individual the same? If not you should try first the k-mean option you have when doing VPC. Click on binning and select K-means. This will find a set of common reference values for each individual from which the confidence bounds are generated. Try and let me know. If I could get the project it would be easier.
